© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 8993
    Prev
    Next
    INDEX
    Explanations

    descriptive phrases related to physical attributes and comparisons

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     tiny
    -0.24
     pequ
    -0.22
     Small
    -0.21
    tiny
    -0.20
     small
    -0.20
     SMALL
    -0.20
    Tiny
    -0.19
    Small
    -0.19
     Tiny
    -0.19
     peque
    -0.19
    POSITIVE LOGITS
     tall
    0.37
     towering
    0.32
     giant
    0.30
     tallest
    0.29
     taller
    0.27
     large
    0.26
     giants
    0.26
     Tall
    0.25
     big
    0.24
     Giant
    0.24
    Activations Density 0.016%

    No Known Activations